The flashcards below were created by user
on FreezingBlue Flashcards.
What is a frequency distribution?
- It summarizes the data into a more useful form.
- Condenses the data.
- Mostly seen as Table 1 in clinical trials
How do you calculate the relative frequency?
Pick a category and divide that by the total # of results to get a decimal.
What graph do u use to graph a frequency distribution?
- Histogram - no spaces between the bars
- Excel can't make histogram
For the histogram what goes on the the horizontal axis?
Class boundaries, categories, etc
What goes on the vertical axis of a histogram?
What levels of measurement are used in histograms?
Ordinal, Interval, and ratio
What are bar/pie graphs used for?
Used for nominal and ordinal or both
In a bar or pie chart what shows the frequency or percentage for each category?
- Bar Chart - height of bar
- Pie Chart - pie slice
- The bigger the higher frequency
Nominal and Ordinal together?
Interval and Ratio together?
What is a true bar chart?
- The bars go left to right not up to down.
- Has Spaces
What do you use to graph for a histogram, bar, and pie chart?
- Histogram - categorie + frequency
- Bar - categorie + percentage
- Pie - categorie + percentage
What is the mean?
Average of the data
Technically what is mean, mode, median?
- This is your First Name
What is your last name?
What is your middle name?
What is the population mean?
What is the most common measure of central tendency?
Mean - affected by outliers
What is mean?
- The fulcrum - from jeoprady
- Its continuous data - interval and ratio
What type of mean is used in stats?
What is the median?
Robust measure of central tendency because not affected by outliers
What must be done to get the median?
Arrange it in numerical order first
What is mode?
- Value occurs most often
- Not effected by outlier
- For numerical or categorical data
How many modes can there be?
None, 1, or multiple (use commas arrange small to big)
What is the best for measurement?
- Mean only if no outliers
- Then Median
- Only mode if categorical data
What quartiles are measures of noncentral location?
Q1 and Q3
What quartile gives you the median or measure of central tendency?
What measures variation?
- Range (large - small)
- Interquartile Range
- Standard Deviation (most important)
Does range provide a good value?
No, it ignores the way in which data are distributed
What is interquartile range (midspread)?
- Measure of variation
- Difference between third and first quartile
- Not affected by extreme values
- Goes with the median
What is variance?
- Measure of Variation
- Shows variation about the mean
What is standard deviation?
- Measure of Variation
- Shows variation of the mean
- Sq root of the variance
What is standard error?
- To understand variation within many samples
- Will always be smaller than st. dev.
What is the Empirical Rule?
- Data is bell-shaped (symmetric) for the population or LARGE samples
- 1. M =- 1o contains about 68% of the values
- 2. 2o = 95%
- 3. 3o = 99.7%
What are the 2 types of distributions?
- Variable of interest distribution (Normal/Skewed)
- Probablility Distribution (Normal/Non-Normal)
What is the central limit theorem?
If the sample size gets large enough the sampling distribution becaomes almost normal regardless of shape of population because your getting closer and closer to to the population size
What is normal distribution?
- Bell Shaped
- Mean, mode, median all =
- Location determined by mean
- Spread determined by st dev.
- Has infinite theoretical range: all - to + # (infinity)
What is the standard normal?
Converting the normal distribution into Z units
What does the t value depends on?
- Degrees of Freedom
- d.f = n-1
Whats the difference between t-distributions and normal distribution graphs?
- Same bell shape and symmetric but fatter tails - means triple M = 0 but std dev depends on df
- Smaller the DF the flatter the graph
How do you know how many tails in the T test?
- Null Hypothesis is of Equality (=) - 2 tail test
- Non inferiority (< , >) - 1 tail
What is the difference between F-distributions and T-distributions?
- DF - 2 for F
- One tailed only for F
- Skewed to right for F
What can you use Chi square test for?
- Small Sample Size continuous data
- Badly skewed data
- Nominal Data
- Ordinal Data